SPARK-1284: Fix improper use of SimpleDateFormat

`SimpleDateFormat` is not thread-safe. Some places use the same SimpleDateFormat object without safeguard in the multiple threads. It will cause that the Web UI displays improper date.

This PR creates a new `SimpleDateFormat` every time when it's necessary. Another solution is using `ThreadLocal` to store a `SimpleDateFormat` in each thread. If this PR impacts the performance, I can change to the latter one.

Author: zsxwing <[email protected]>

Closes #179 from zsxwing/SPARK-1278 and squashes the following commits:

21fabd3 [zsxwing] SPARK-1278: Fix improper use of SimpleDateFormat

Conflicts:

	core/src/main/scala/org/apache/spark/scheduler/JobLogger.scala
	core/src/main/scala/org/apache/spark/ui/jobs/JobProgressUI.scala
	core/src/main/scala/org/apache/spark/ui/jobs/StagePage.scala
	core/src/main/scala/org/apache/spark/ui/jobs/StageTable.scala
	core/src/main/scala/org/apache/spark/util/FileLogger.scala

Author: zsxwing

core/src/main/scala/org/apache/spark/deploy/master/Master.scala

@@ -45,7 +45,7 @@ private[spark] class Master(host: String, port: Int, webUiPort: Int) extends Act
 
   val conf = new SparkConf
 
-  val DATE_FORMAT = new SimpleDateFormat("yyyyMMddHHmmss")  // For application IDs
+  def createDateFormat = new SimpleDateFormat("yyyyMMddHHmmss")  // For application IDs
   val WORKER_TIMEOUT = conf.getLong("spark.worker.timeout", 60) * 1000
   val RETAINED_APPLICATIONS = conf.getInt("spark.deploy.retainedApplications", 200)
   val REAPER_ITERATIONS = conf.getInt("spark.dead.worker.persistence", 15)
@@ -621,7 +621,7 @@ private[spark] class Master(host: String, port: Int, webUiPort: Int) extends Act
 
   /** Generate a new app ID given a app's submission date */
   def newApplicationId(submitDate: Date): String = {
-    val appId = "app-%s-%04d".format(DATE_FORMAT.format(submitDate), nextAppNumber)
+    val appId = "app-%s-%04d".format(createDateFormat.format(submitDate), nextAppNumber)
     nextAppNumber += 1
     appId
   }
@@ -644,7 +644,7 @@ private[spark] class Master(host: String, port: Int, webUiPort: Int) extends Act
   }
 
   def newDriverId(submitDate: Date): String = {
-    val appId = "driver-%s-%04d".format(DATE_FORMAT.format(submitDate), nextDriverNumber)
+    val appId = "driver-%s-%04d".format(createDateFormat.format(submitDate), nextDriverNumber)
     nextDriverNumber += 1
     appId
   }

core/src/main/scala/org/apache/spark/deploy/master/ui/IndexPage.scala

@@ -25,10 +25,10 @@ import akka.pattern.ask
 import javax.servlet.http.HttpServletRequest
 import net.liftweb.json.JsonAST.JValue
 
-import org.apache.spark.deploy.{DeployWebUI, JsonProtocol}
+import org.apache.spark.deploy.JsonProtocol
 import org.apache.spark.deploy.DeployMessages.{MasterStateResponse, RequestMasterState}
 import org.apache.spark.deploy.master.{ApplicationInfo, DriverInfo, WorkerInfo}
-import org.apache.spark.ui.UIUtils
+import org.apache.spark.ui.{WebUI, UIUtils}
 import org.apache.spark.util.Utils
 
 private[spark] class IndexPage(parent: MasterWebUI) {
@@ -164,10 +164,10 @@ private[spark] class IndexPage(parent: MasterWebUI) {
       <td sorttable_customkey={app.desc.memoryPerSlave.toString}>
         {Utils.megabytesToString(app.desc.memoryPerSlave)}
       </td>
-      <td>{DeployWebUI.formatDate(app.submitDate)}</td>
+      <td>{WebUI.formatDate(app.submitDate)}</td>
       <td>{app.desc.user}</td>
       <td>{app.state.toString}</td>
-      <td>{DeployWebUI.formatDuration(app.duration)}</td>
+      <td>{WebUI.formatDuration(app.duration)}</td>
     </tr>
   }
 

core/src/main/scala/org/apache/spark/deploy/worker/Worker.scala

@@ -56,7 +56,7 @@ private[spark] class Worker(
   Utils.checkHost(host, "Expected hostname")
   assert (port > 0)
 
-  val DATE_FORMAT = new SimpleDateFormat("yyyyMMddHHmmss")  // For worker and executor IDs
+  def createDateFormat = new SimpleDateFormat("yyyyMMddHHmmss")  // For worker and executor IDs
 
   // Send a heartbeat every (heartbeat timeout) / 4 milliseconds
   val HEARTBEAT_MILLIS = conf.getLong("spark.worker.timeout", 60) * 1000 / 4
@@ -309,7 +309,7 @@ private[spark] class Worker(
   }
 
   def generateWorkerId(): String = {
-    "worker-%s-%s-%d".format(DATE_FORMAT.format(new Date), host, port)
+    "worker-%s-%s-%d".format(createDateFormat.format(new Date), host, port)
   }
 
   override def postStop() {

core/src/main/scala/org/apache/spark/scheduler/JobLogger.scala

@@ -53,6 +53,9 @@ class JobLogger(val user: String, val logDirName: String)
   private val jobIDToPrintWriter = new HashMap[Int, PrintWriter]
   private val stageIDToJobID = new HashMap[Int, Int]
   private val jobIDToStages = new HashMap[Int, ListBuffer[Stage]]
+  private val dateFormat = new ThreadLocal[SimpleDateFormat]() {
+    override def initialValue() = new SimpleDateFormat("yyyy/MM/dd HH:mm:ss")
+  }
   private val DATE_FORMAT = new SimpleDateFormat("yyyy/MM/dd HH:mm:ss")
   private val eventQueue = new LinkedBlockingQueue[SparkListenerEvents]
 
@@ -116,7 +119,7 @@ class JobLogger(val user: String, val logDirName: String)
     var writeInfo = info
     if (withTime) {
       val date = new Date(System.currentTimeMillis())
-      writeInfo = DATE_FORMAT.format(date) + ": " +info
+      writeInfo = dateFormat.get.format(date) + ": " + info
     }
     jobIDToPrintWriter.get(jobID).foreach(_.println(writeInfo))
   }

core/src/main/scala/org/apache/spark/deploy/WebUI.scala renamed to core/src/main/scala/org/apache/spark/ui/WebUI.scala

@@ -15,20 +15,23 @@
  * limitations under the License.
  */
 
-package org.apache.spark.deploy
+package org.apache.spark.ui
 
 import java.text.SimpleDateFormat
 import java.util.Date
 
 /**
  * Utilities used throughout the web UI.
  */
-private[spark] object DeployWebUI {
-  val DATE_FORMAT = new SimpleDateFormat("yyyy/MM/dd HH:mm:ss")
+private[spark] object WebUI {
+  // SimpleDateFormat is not thread-safe. Don't expose it to avoid improper use.
+  private val dateFormat = new ThreadLocal[SimpleDateFormat]() {
+    override def initialValue() = new SimpleDateFormat("yyyy/MM/dd HH:mm:ss")
+  }
 
-  def formatDate(date: Date): String = DATE_FORMAT.format(date)
+  def formatDate(date: Date): String = dateFormat.get.format(date)
 
-  def formatDate(timestamp: Long): String = DATE_FORMAT.format(new Date(timestamp))
+  def formatDate(timestamp: Long): String = dateFormat.get.format(new Date(timestamp))
 
   def formatDuration(milliseconds: Long): String = {
     val seconds = milliseconds.toDouble / 1000

core/src/main/scala/org/apache/spark/ui/jobs/ExecutorTable.scala

@@ -27,7 +27,6 @@ import scala.collection.mutable
 private[spark] class ExecutorTable(val parent: JobProgressUI, val stageId: Int) {
 
   val listener = parent.listener
-  val dateFmt = parent.dateFmt
   val isFairScheduler = listener.sc.getSchedulingMode == SchedulingMode.FAIR
 
   def toNodeSeq(): Seq[Node] = {

core/src/main/scala/org/apache/spark/ui/jobs/JobProgressUI.scala

@@ -40,7 +40,6 @@ import org.apache.spark.util.Utils
 private[spark] class JobProgressUI(val sc: SparkContext) {
   private var _listener: Option[JobProgressListener] = None
   def listener = _listener.get
-  val dateFmt = new SimpleDateFormat("yyyy/MM/dd HH:mm:ss")
 
   private val indexPage = new IndexPage(this)
   private val stagePage = new StagePage(this)

core/src/main/scala/org/apache/spark/ui/jobs/StagePage.scala

@@ -26,14 +26,14 @@ import scala.xml.Node
 import org.apache.spark.{ExceptionFailure}
 import org.apache.spark.executor.TaskMetrics
 import org.apache.spark.ui.UIUtils._
+import org.apache.spark.ui.WebUI
 import org.apache.spark.ui.Page._
 import org.apache.spark.util.{Utils, Distribution}
 import org.apache.spark.scheduler.TaskInfo
 
 /** Page showing statistics and task list for a given stage */
 private[spark] class StagePage(parent: JobProgressUI) {
   def listener = parent.listener
-  val dateFmt = parent.dateFmt
 
   def render(request: HttpServletRequest): Seq[Node] = {
     listener.synchronized {
@@ -248,7 +248,7 @@ private[spark] class StagePage(parent: JobProgressUI) {
       <td>{info.status}</td>
       <td>{info.taskLocality}</td>
       <td>{info.host}</td>
-      <td>{dateFmt.format(new Date(info.launchTime))}</td>
+      <td>{WebUI.formatDate(new Date(info.launchTime))}</td>
       <td sorttable_customkey={duration.toString}>
         {formatDuration}
       </td>

core/src/main/scala/org/apache/spark/ui/jobs/StageTable.scala

@@ -23,15 +23,14 @@ import scala.xml.Node
 import scala.collection.mutable.HashSet
 
 import org.apache.spark.scheduler.{SchedulingMode, StageInfo, TaskInfo}
-import org.apache.spark.ui.UIUtils
+import org.apache.spark.ui.{WebUI, UIUtils}
 import org.apache.spark.util.Utils
 
 
 /** Page showing list of all ongoing and recently finished stages */
 private[spark] class StageTable(val stages: Seq[StageInfo], val parent: JobProgressUI) {
 
   val listener = parent.listener
-  val dateFmt = parent.dateFmt
   val isFairScheduler = listener.sc.getSchedulingMode == SchedulingMode.FAIR
 
   def toNodeSeq(): Seq[Node] = {
@@ -75,7 +74,7 @@ private[spark] class StageTable(val stages: Seq[StageInfo], val parent: JobProgr
 
   private def stageRow(s: StageInfo): Seq[Node] = {
     val submissionTime = s.submissionTime match {
-      case Some(t) => dateFmt.format(new Date(t))
+      case Some(t) => WebUI.formatDate(new Date(t))
       case None => "Unknown"
     }